Breast cancers display great phenotypic and molecular diversity with subtypes showing striking differences in prognosis and drug responses. However, little is currently known about the parameters contributing to this heterogeneity. Our laboratory has addressed this issue with a long-term program to develop an experimentally tractable human mammary epithelial cell (HMEC) culture system for investigating multi-step human breast carcinogenesis. Understanding the factors that propel type-specific breast cancers may facilitate better clinical approaches for treatment and prevention. Hypothesis/Objective: We hypothesize that breast cancer heterogeneity derives from molecularly distinct cells of origin and progression pathways. Our objectives are to systematically generate a panel of transformed HMEC from 8 different donors that reflect the in vivo spectrum of breast cancer phenotypes, to molecularly characterize the isogenic cells at different stages of transformation to better understand the factors that propel cancer progression, and to make these new reagents available to the scientific community. Specific Aim 1. Develop a panel of in vitro transformed HMEC lines that reflect the diverse progression pathways and subtypes seen in human breast cancer in vivo. The variables to be employed include: specimen age (<30 yr vs >55 yr); target population (FACS-enriched progenitor, myoepithelial, and luminal cells); agents to bypass stasis (shRNA to p16, cyclin D1 overexpression, shRNA to RB); agents to confer malignancy (mutant HER2, mutant PIK3CA). Specific Aim 2. Characterize resultant HMEC lines for properties associated with lineage and cancer subtypes. Properties to be examined include lineage markers (FACS, IF, IHC), gene transcript profiles, and AIG. This information may identify subtype-specific cells of origin and molecular alterations. Although over 80% of breast cancers are classified as luminal, the in vitro transformed cell lines available are almost exclusively basal. As a result, the relevance of experiments based on e.g. MCF10A may be limited to a small proportion of actual breast cancer cases. By providing experimentally tractable in vitro models of breast cancer diversity we hope to uncover valuable etiologic information and facilitate development of improved methods for prevention, prognosis, and type specific clinical interventions. Our comprehensive system would be suitable for high-throughput drug discovery studies. Disseminating our resources to the overall cancer research community can amplify their usefulness in efforts to reduce the burden of breast cancer.